AI Disclosure Statement
An AI disclosure statement is a short explanation of how AI was used in creating a work. This page explains how that differs from a structured provenance disclosure, which provides a fuller record of the creation process and is the type of document Provenance Disclosure is designed to generate.
In other words, a statement is usually the lightweight public summary. A provenance disclosure is the more formal record behind it.
What a statement is and what it is not
A statement is usually brief. It may appear in a publication, release note, policy page, submission, or internal governance document as a concise transparency note.
It is not usually a full evidentiary record. It does not, by itself, document every tool, every workflow step, or the complete basis for the representation being made.
That distinction matters. A short statement can be useful for communication, while a fuller disclosure record is better when accountability, reviewability, or future reference matter.
What it usually answers
A useful AI disclosure statement usually addresses three points:
- Whether AI tools were used at all
- What role those tools played in the workflow
- Whether a human reviewed and approved the final result
The goal is not to make every possible claim. The goal is to communicate the role of automation clearly and honestly.
Where a statement usually comes from
An AI disclosure statement normally comes from the publisher, responsible team, author, organization, or reviewer releasing the work. In other words, it is a declared representation by the party taking responsibility for the output.
That is different from a classifier result, outside guess, or detection score. A statement is not an inference about the work. It is a description offered by the responsible party.
When organizations use one
Organizations commonly publish AI disclosure statements in product documentation, research submissions, editorial workflows, procurement reviews, and internal governance materials.
In those contexts, the statement functions as a concise summary of process rather than a complete record.
How it differs from a provenance disclosure
An AI disclosure statement is usually short and high level. A provenance disclosure is more structured. It can document the role of AI systems, the role of human reviewers, the basis for the declaration, and the boundaries of what is and is not being claimed.
So the difference is not just length. The difference is function:
- A statement communicates.
- A disclosure records.
If you only need a short transparency note, a statement may be enough. If you need a durable, attributable, and reviewable record, a provenance disclosure is the stronger format.
Example statement language
Artificial intelligence tools were used to assist with drafting and organization of this material. A human reviewer edited, evaluated, and approved the final content before release.
This is the kind of language a team might publish alongside a document, release, or public artifact. It is useful, but it is still only a summary.
How this relates to Provenance Disclosure
Provenance Disclosure is not an AI detector and not just a statement generator. It is designed to generate a structured provenance disclosure: a more formal record of how a work was created, what role automation played, and who took responsibility for the final result.
In practice, a team might publish a short AI disclosure statement publicly while also generating a provenance disclosure internally or for counterparties who need a fuller record.
Generate a structured disclosure
If you need something more durable than a short statement, you can generate a structured provenance disclosure describing the role of automation, human review, and the basis for the declaration.